feat(collection-seeder): add influenza resistance mutation collections

[fhennig]

resolves #1272

Summary

Adds resistance mutations for all influenza viruses.

Screenshot

Screen.Recording.2026-07-14.at.15.10.05.mov

PR Checklist

• All necessary documentation has been adapted.
• The implemented feature is covered by an appropriate test.

[vercel]

The latest updates on your projects. Learn more about Vercel for GitHub.

Project	Deployment	Actions	Updated (UTC)
dashboards	Ready	Preview, Comment	Jul 15, 2026 12:00pm

[Copilot]

Pull request overview

Adds a new collection-seeding source that pulls influenza antiviral resistance mutations from the BU-ISCIII flu_resistance repository and generates seeded collections for the influenza dashboards (H1N1pdm, H3N2, H5N1, and B/Victoria), resolving #1272.

Changes:

• Introduces influenza_resistance_mutations seeding sources that fetch NA- and PA-inhibitor mutation datasets and build collections/variants per subtype.
• Registers the new influenza sources in the seeder source registry so they run in the default seed flow.
• Extends the shared Variant typed model to allow an optional description, and adds a dedicated test suite for the influenza sources and parsing.

Reviewed changes

Copilot reviewed 4 out of 4 changed files in this pull request and generated 3 comments.

File	Description
collection-seeding/sources/influenza_resistance_mutations.py	New influenza resistance mutation sources and JSON fetching/parsing to produce collections.
collection-seeding/sources/registry.py	Registers the new influenza sources in ALL_SOURCES.
collection-seeding/models.py	Makes Variant.description optional via NotRequired.
collection-seeding/tests/test_influenza_resistance_mutations.py	Adds unit tests for parsing and for collection/variant generation across influenza subtypes.

[Copilot]

Pull request overview

Copilot reviewed 4 out of 4 changed files in this pull request and generated no new comments.

[fengelniederhammer]

AI: 🔴 bug: @lru_cache persists for the whole pytest process, which breaks test isolation. The first fetch test (test_h1n1_organism) populates the cache; every later @responses.activate test then registers NA/PA mocks that never fire, and responses (0.26.2, default assert_all_requests_are_fired=True) fails any test with unfired mocks — so ~15 tests go red. It also leaks fixture data across tests. Clear the cache between tests, e.g. add a conftest.py (or module-level fixture) with:

import pytest
from sources.influenza_resistance_mutations import _fetch_json

@pytest.fixture(autouse=True)
def _clear_fetch_cache():
    _fetch_json.cache_clear()
    yield
    _fetch_json.cache_clear()

(Could not run the suite locally — assessment from the responses default + call ordering; please confirm on CI.)

[fengelniederhammer]

(Not sure whether that's true - in any case I think it's not that important)